Saves the optimizer and network states during training. The final network and optimizer are always stored.
Details
Saving the learner itself in the callback with a trained model is impossible, as the model slot is set after the last callback step is executed.
See also
Other Callback:
TorchCallback,
as_torch_callback(),
as_torch_callbacks(),
callback_set(),
mlr3torch_callbacks,
mlr_callback_set,
mlr_callback_set.progress,
mlr_context_torch,
t_clbk(),
torch_callback()
Super class
mlr3torch::CallbackSet -> CallbackSetCheckpoint
Methods
Method new()
Creates a new instance of this R6 class.
Usage
CallbackSetCheckpoint$new(path, freq, freq_type = "epoch")Arguments
path(
character(1))
The path to a folder where the models are saved.freq(
integer(1))
The frequency how often the model is saved. Frequency is either per step or epoch, which can be configured through thefreq_typeparameter.freq_type(
character(1))
Can be be either"epoch"(default) or"step".
Method on_epoch_end()
Saves the network and optimizer state dict.
Does nothing if freq_type or freq are not met.
Method on_batch_end()
Saves the selected objects defined in save.
Does nothing if freq_type or freq are not met.